High Expression of CDCA7 Predicts Tumor Progression and Poor Prognosis in Human Colorectal Cancer

Molecular Medicine REPORTS 22: 57-66, 2020

High expression of CDCA7 predicts tumor progression and poor prognosis in human colorectal cancer

SIMAN LI1*, JIEAN HUANG2*, MENGBIN QIN2, JINXIU ZHANG2 and CUN LIAO3

1Department of Gastroenterology, The First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi 530021; 2Department of Gastroenterology, The Second Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi 530007; 3Department of Colorectal‑Anal Surgery, The First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi 530021, P.R. China

Received April 13, 2019; Accepted February 7, 2020

DOI: 10.3892/mmr.2020.11089

Abstract. Colorectal cancer (CRC) is one of the most fatal Introduction types of cancer worldwide. This study aimed to determine the predictive and prognostic values of cell division cycle Colorectal cancer (CRC) is recognized as one of the most associated protein 7 (CDCA7) in CRC. Firstly, the relationship common and deadly cancers worldwide (1). It occupies an between CDCA7 and CRC was assessed through bioinfor- increasingly important position in the spectrum of cancer, and matics analysis. Subsequently, CDCA7 expression levels were represents the leading cause of cancer‑related mortality world- detected in various CRC cell lines, as well as 15 fresh human wide (2,3). Due to the lack of specific clinical symptoms and CRC tissues and their paired adjacent normal colorectal tissues early screening tests, a substantial number of patients with CRC using reverse transcription‑quantitative PCR and western blot- are first diagnosed at an advanced stage with metastasis (4). ting. Additionally, immunohistochemical staining was used Furthermore, even if a tumor is resected, CRC recurrence and to determine the levels of CDCA7 in 104 CRC tissues and metastasis may occur after surgery; therefore, the prognosis their paired adjacent normal colorectal tissues. The present of CRC remains poor (5). Thus, it is of great importance to study revealed that CDCA7 expression was upregulated in understand the molecular mechanisms underlying CRC devel- CRC tissues and cell lines. The positive expression rates of opment and to identify novel tumor‑associated biomarkers. CDCA7 in normal and CRC tissues were 26.92 and 75.96%, Bioinformatics analysis provides a deeper and more respectively. The intensities of CDCA7 immunostaining were comprehensive understanding of targeted genes, and allows significantly associated with CRC invasion depth, lymph node researchers to identify functional genes for further analysis. metastasis, tumor‑node‑metastasis stage and distant metas- The Cancer Genome Atlas (TCGA) generates comprehensive, tasis. However, no significant differences in sex, age, tumor multi‑dimensional maps of the key genomic changes in cancer, size and CRC differentiation were found between high and and provides researchers with a variety of genetic informa- low CDCA7 expression groups. Furthermore, patients with tion on different types of cancer (6). Kyoto Encyclopedia of low CDCA7 expression exhibited a greater overall survival Genes and Genomes (KEGG) (7) and Gene Ontology (GO) (8) rate of CRC compared to those with high CDCA7 expression. analyses have enabled the identification of metabolic signaling The findings of this study indicated that CDCA7 may serve a pathways and major biological functions of differentially significant role in CRC prognosis and progression, and may expressed genes. Therefore, TCGA, data collection, and be considered a novel biomarker for the prediction of patient GO and KEGG pathway enrichment analyses were performed survival after colectomy. in this study. Cell division cycle associated protein 7 (CDCA7) is located on chromosome 2q31 and encodes a nuclear protein containing 371 amino acids (9). CDCA7 is a cell division cycle‑associated protein that was first discovered in Myc‑transfected fibro- Correspondence to: Dr Cun Liao, Department of Colorectal‑anal blasts, which is commonly overexpressed in various types of Surgery, The First Affiliated Hospital of Guangxi Medical human cancer (10). CDCA7 is periodically expressed during University, 6 Shuangyong Road, Nanning, Guangxi 530021, the human cell cycle, with the highest expression level found P. R. Ch i na E‑mail: [email protected] in the G1 to S phase (11). Deregulation of cell cycle proteins often leads to the increased risk of tumor occurrence (12,13). *Contributed equally Previous reports have demonstrated that CDCA7 is a c‑Myc‑responsive gene that participates in cancer transfor- Key words: cell division cycle associated protein 7, colorectal mation and tumorigenesis (9). Thus far, several studies have cancer, progression, prognosis, bioinformatics analysis reported that CDCA7 is overexpressed in different types of cancer, including lymphoma, ovarian cancer, retinoblastoma, breast cancer, acute myeloid leukemia and esophageal 58 LI et al: HIGH EXPRESSION OF CDCA7 PREDICTS TUMOR PROGRESSION AND POOR PROGNOSIS IN CRC cancer (10,14‑18). These findings provide evidence that there patterns. The integrated score based on the two predicted may be an association between CDCA7 and tumor develop- scores in the TRANSFAC® database (http://genexplain. ment. However, to the best of our knowledge, the relationship com/transfac/#section3) were assigned as the final predicted between CDCA7 and the development of colorectal disease score. Binding regions of the predicted transcription factors has not yet been reported on. Therefore, the clinical relevance were annotated based on the DNA methylation and single and underlying mechanisms of CDCA7 in the occurrence and nucleotide polymorphism (SNP) information in COSMIC (19) development of CRC remain largely unclear. The present study and dbSNP (https://www.ncbi.nlm.nih.gov/snp/) databases, aimed to investigate the expression levels and clinical role of respectively. According to the predicted scores of the CDCA7 in CRC. TRANSFAC® database and whether the corresponding comment information existed in the COSMIC database and Materials and methods the dbSNP database, the recommendation degree was integrated. The recommendation degree was used to represent the Bioinformatics analysis research value of the predicted transcription factor; the higher Collection of CDCA7 expression data from TCGA. the recommendation degree, the better. Results with a high microRNA (miR/miRNA) expression matrix was retrieved recommendation degree were selected for use in the present from TCGA (https://portal.gdc.cancer.gov). RNA‑seq data study. from 647 CRC tissues (cancer group) and 51 normal adjacent tissues (normal group) were downloaded from TCGA Gene expression in tumors. According to TCGA barcodes, database (dataset nos. TCGA‑COAD and TCGA‑READ). all samples were classified into normal and tumor tissues. Differentially expressed levels of CDCA7 between the cancer Subsequently, the average value and standard deviation of group and normal group were identified using the Ballgown all samples were determined based on the expression values (https://github.com/alyssafrazee/ballgown) package with the in the normalized‑results file of each sample provided by following criteria: P<0.05 and fold‑change >2. RNA‑SEQv2 data. Finally, the target gene expression in human tumors was interpreted. Gene functions and pathways. KEGG (http://www.genome. jp/kegg/pathway.html) and GO (http://geneontology.org/) Tissue samples. CRC tissues and the paired adjacent were used for the enrichment analysis of dysregulated genes, non‑tumor tissues (n=104) in the form of pathological slices in order to determine the changes in functions and pathways were collected from patients (sex, 61 males and 43 females; associated with CRC. The conditional criterion was P<0.05. age, 26‑82 years; average age, 56.42±12.87 years) who under- went surgical operation at the Department of Colorectal‑anal Research hotspots. According to the gene annotation in the Surgery, The First Affiliated Hospital of Guangxi Medical database, genetic studies were evaluated based on five different University (Guangxi, China) between October 2012 and perspectives: i) Relevant literature of CDCA7 (total number of November 2013. In addition, CRC tissues were collected from articles reported); ii) pathways (number of pathways that CDCA7 15 patients (sex, 9 males and 6 females; age, 47‑75 years; was involved in according to the KEGG database); iii) functions average age, 60.33±8.83 years) with pathological confirmation (number of functions that involved the CDCA7 gene according of the diagnosis in November 2017. All tissue specimens were to the GO database); iv) validated targeted miRNAs (number collected within 30 min after surgical resection. The 15 pairs of miRNAs associated with CDCA7 that have been published of CRC tissues and normal adjacent tissues were immediately in literature); and v) diseases (number of diseases associated snap‑frozen in liquid nitrogen, and then stored at ‑80˚C until with CDCA7 that have been published in the literature), in order protein and RNA extraction. The survival status of patients to identify research hotspots of CDCA7 through the KEGG with CRC was verified by telephone interview. None of the and GO databases, and published literature. A radar chart was patients received pre‑operative chemoradiotherapy or radio- produced using Microsoft Office 2013 software to demonstrate therapy. Clinicopathological characteristics of the patients the research hotspots of CDCA7 (Fig. 1). were assessed by reviewing patients' medical records and pathology reports. Disease association. The frequency of the gene reported The ethical approval for this study was obtained from the in different diseases was mined from the PubMed database Protection of Human Ethics Committee of The First Affiliated (https://www.ncbi.nlm.nih.gov/pubmed/) according to the MeSH Hospital of Guangxi Medical University. All specimens were disease classification (https://www.ncbi.nlm.nih.gov/mesh). The made anonymous and handled in accordance with the legal top 20 diseases associated with the target gene were sought out. and ethical standards for protecting human rights.

Regulatory relationship. The regulatory relationship among Cell lines and cell culture. Human normal colonic epithelial genes was examined based on four different perspectives: cell line NCM460 and human CRC cell line SW620 were Gene‑related transcription factors, miRNAs, long non‑coding obtained from Wuhan Boster Biological Technology, Ltd. (lnc)RNAs, as well as upstream and downstream gene Other human CRC cell lines, HT29, HCT116 and RKO, sequences, via published literature and the databases of were purchased from Shanghai R&S Biotechnology Co., PubMed, MeSH and KEGG. Ltd. These cell lines were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with FBS Transcription factor prediction. Transcription factors (100 ml/l; Shanghai ExCell Biology, Inc.) and maintained were predicted according to the positional binding in 5% CO2 at 37˚C. Mycoplasma testing was then performed Molecular Medicine REPORTS 22: 57-66, 2020 59

Figure 1. Research hotspots of cell division cycle associated protein 7. GO, Gene Ontology; miRNA, microRNA. The gray area represents the average number of research fever of all genes, while the green area represents the number of research fever of CDCA7. The dots represent the reported amount of CDCA7 in the field.

on all the cultured cell lines. The four human CRC cell lines temperature, the sections were stained with DAB reagent for were referenced to the NCM460 cell line, which acted as the 6 min, counterstained with hematoxylin for 30 sec and sealed control group. with neutral gum. Significant yellow or yellow‑brown granules appeared in the cell membrane, cytoplasm or nucleus, which Immunohistochemistry. Immunohistochemistry was was regarded as positive streptavidin peroxidase (SP) staining. performed using SPlink Detection kits (cat. no. SP‑9000; Semi‑quantitative analysis was conducted by integrating the OriGene Technologies, Inc.), according to the manufacturer's staining intensity and the percentage of positive cells. The protocol. Tissue specimens were fixed in 10% formalin for 48 h stained tissue sections were observed under an Olympus optical at room temperature. After dehydration and paraffin embed- microscope (magnification, x400). The values of staining ding, all the tissue specimens were cut into 4‑µm sections. index (0‑12) were calculated by multiplying the two scores Specifically, at room temperature, colon tissue sections were obtained (intensity and percentage). The staining intensities routinely dewaxed with xylene and dehydrated in a descending of the positive cells were categorized as follows: 0 point (nega- alcohol series, prior to being incubated in 0.01 M citrate buffer tive); 1 point (weak); 2 points (moderate); and 3 points (strong). at 100˚C for 6 min for antigen retrieval. Following being A total of five different fields were randomly observed for cooled down, the sections were washed three times in PBS each slice, and the percentages of positively‑stained cells for 5 min each and the sections were then incubated with were categorized as follows: 0 points (<5%); 1 point (5‑25%);

3% H2O2 at room temperature for 12 min. Subsequently, after 2 points (26‑50%); 3 points (51‑75%); and 4 points (>75%). being washed thrice, sections were incubated with goat serum Final staining scores of 0‑7 indicated low CDCA7 expression, (provided by the kit) at room temperature for 12 min. Following whereas staining scores of 8‑12 represented high CDCA7 being blocked, the tissue sections were incubated with CDCA7 expression. Each slide was independently examined by polyclonal primary antibody (1:100; cat. no. PA5‑52165; two experienced pathologists. Invitrogen; Thermo Fisher Scientific, Inc.) at 4˚C overnight. The following day, the incubated tissue sections were washed RNA extraction, cDNA synthesis and reverse transcrip‑ three times with PBS after being left at room temperature for tion‑quantitative PCR (RT‑qPCR). Total RNA was ≥30 min. The primary antibody was replaced with PBS as a extracted from 30 samples of fresh CRC tissues and adja- negative control. Subsequently, the tissue sections were incu- cent normal colorectal tissues, as well as CRC cell lines bated with a biotin‑conjugated secondary antibody (provided using Eastep® Super Total RNA Extraction kit (Promega by the kit) at room temperature for 12 min. Subsequently, after Corporation), according to the manufacturer's protocol. being washed with PBS three times, the tissue sections were Subsequently, cDNA was synthesized by the GoScript™ incubated with horseradish peroxidase‑conjugated strepto- Reverse Transcription Mix, Random Primers kit (Promega mycin avidin (provided by the kit) at room temperature for Corporation). The following RT temperature protocol was 12 min. The sections were then repeatedly immersed 3 times used: 5 min at 25˚C, 60 min at 42˚C and 15 min at 70˚C, in PBS for 5 min each. After drying the sections at room prior to being maintained at 4˚C. RT‑qPCR was conducted 60 LI et al: HIGH EXPRESSION OF CDCA7 PREDICTS TUMOR PROGRESSION AND POOR PROGNOSIS IN CRC

Table I. Primer sequences used for reverse transcription‑quan- Western blot analysis. Total protein was isolated from CRC titative PCR. tissues, normal colorectal tissues and CRC cell lines using RIPA lysis buffer containing 1% phosphatase inhibitor and Gene Primer sequences (5'→3') 1% protease inhibitor. After incubation at 0˚C for 30 min, the lysed cells were washed twice with ice‑cold PBS, and then GAPDH F: GCACCGTCAAGGCTGAGAAC centrifuged at 12,000 x g and 4˚C for 15 min to eliminate R: TGGTGAAGACGCCAGTGGA potential cell debris. Total protein concentration was deter- CDCA7 F: CCAGGCTCCGACTCACAATCAAG mined by a bicinchoninic acid assay (Beyotime Institute of R: GTACTTATCCTCTTCCTCCTCCTCCTC Biotechnology). Subsequently, the quantified proteins (50 µg) were denatured and subjected to SDS‑PAGE on 12% gels, CDCA7, cell division cycle associated protein 7; F, forward; R, and transferred onto 0.22 µm PVDF membranes. The reverse. membranes were then blocked with non‑fat milk (5%) for 1 h at 37˚C. After washing with Tris‑buffered saline containing 5% Tween‑20 (TBST) three times, the membranes were incubated with rabbit polyclonal CDCA7 antibody (1:250; Table II. Gene function and pathway analysis of CDCA7. cat. no. PA5‑52165; Invitrogen; Thermo Fisher Scientific, Inc.) and mouse monoclonal GAPDH antibody (1:2,000; Gene ID GO function cat. no. 60004‑1‑lg; ProteinTech Group, Inc.) antibodies. After incubation at 4˚C overnight, the membranes were incubated GO:0006351 regulation, DNA template with IRDye® 680RD goat anti‑rabbit immunoglobulin G (IgG; GO:0006355 Transcriptional regulation, DNA template 1:10,000; cat. no. 925‑68071; LI‑COR Biosciences) and goat GO:0006915 Apoptosis anti‑mouse IgG (1:5,000; cat. no. BA1038; Wuhan Boster GO:0042127 regulates cell proliferation Biological Technology, Ltd.) secondary antibodies, respectively, at room temperature for 1 h. After washing with TBST CDCA7, cell division cycle associated protein 7; GO, Gene Ontology. three times, the stained membranes were visualized using an Odyssey® CLx imaging system (LI‑COR Biosciences). The gray value of each protein was determined by ImageJ version 1.4.3 software (National Institutes of Health). GAPDH Table III. miRNAs and lncRNAs targeted by cell division expression was used as the internal reference. The relative cycle associated protein 7. expression level of CDCA7 was calculated by the ratio of CDCA7 gray value to GAPDH gray value. miRNA lncRNA Statistical analysis. All analyses were carried out using SPSS hsa‑let‑7b‑5p HCP5 Statistics version 20.0 (IBM Corp.). Each experiment was hsa‑miR‑124‑3p TP53TG1 repeated three times. Student's paired t‑test was performed hsa‑miR‑124‑3p KCNQ1OT1 to compare the expression levels of CDCA7 between CRC hsa‑miR‑1254 DKFZP434K028 tissues and adjacent normal colorectal tissues. Expression hsa‑miR‑1254 DKFZP434K028 levels of CDCA7 in CRC cell lines were compared using hsa‑miR‑1271‑3p LINC00341 a one‑way ANOVA and Bonferroni's post hoc test. The hsa‑miR‑1271‑3p BIN3‑IT1 strength of association between CDCA7 expression and CRC 2 hsa‑miR‑193b‑3p DLGAP1‑AS2 clinicopathological features was determined using a χ test or hsa‑miR‑24‑3p LOC90768 Fisher's exact test. Overall survival data were analyzed by the Kaplan‑Meier method and log‑rank test. Data are presented hsa‑miR‑299‑5p BAALC‑AS2 as the means ± SD. P<0.05 was regarded as statistically lncRNA, long non‑coding RNA; miRNA/miR, microRNA. significant.

Results with Applied Biosystems™ 7500 Real‑Time PCR system Bioinformatics prediction of CDCA7 mRNA expression in (Applied Biosystems; Thermo Fisher Scientific, Inc.) using a CRC. Based on the data obtained from TCGA database, ® GoTaq qPCR Master Mix kit (Promega Corporation). qPCR under the criteria of P<0.05 and |log2 fold change|>2, the cycling conditions were as follows: 10 min at 95˚C, followed mRNA expression of CDCA7 was upregulated in CRC tissues by 40 cycles of 15 sec at 95˚C and 60 sec at 60˚C, and a compared to normal tissues (P<0.05), and the fold change dissociation cycle consisting of 15 sec at 95˚C, 15 sec at 60˚C (T/N) was 3.06. and 15 sec at 95˚C (ramping up at 0.2˚C/sec). The expression levels of CDCA7 were measured by 2−ΔΔCq method (20) using GO and KEGG pathway enrichment analysis. The biological GAPDH as a standard reference, and the relative expression functions of CDCA7 in CRC were determined by GO and value was set to 1 under unstimulated conditions. The primer KEGG pathway enrichment analyses. As shown in Table II, sequences (Takara Bio, Inc.) of GAPDH and CDCA7 are according to GO term analysis, it was noted that the most presented in Table I. marked functions of CDCA7 in CRC were apoptosis and Molecular Medicine REPORTS 22: 57-66, 2020 61

Figure 2. Top 20 most commonly reported diseases associated with cell division cycle associated protein 7.

transcription. However, no significant finding was obtained on their positional binding relationship. It was identified that for gene‑involved pathways, as revealed by KEGG pathway upstream‑binding protein 1 (LBP1), zinc finger 5 (ZF5) and analysis. interferon regulatory factor 1 (IRF1) were closely associated with CDCA7 (Fig. 3). Research hotspots of CDCA7. As shown in Fig. 1, a total of 22 articles were reported on CDCA7, in which CDCA7 was Gene expression in tumors. The expression levels of CDCA7 in associated with seven biological GO functions. However, each sample were analyzed by the RNA‑SEQv2 data derived according to the KEGG database, no signaling pathway from TCGA database. The results showed that CDCA7 expres- involving CDCA7 was found. The total number of diseases sion was mainly upregulated in tumor tissues compared to associated with CDCA7 was 31, and 22 different targeted adjacent normal tissues (Fig. 4). miRNAs had been validated. Only 2 (disease and miRNA) of the 5 indicators of CDCA7 were higher than average. Thus, High expression of CDCA7 in CRC tissues. Immunohisto- the lack of findings surrounding CDCA7 indicated that little chemistry SP staining revealed that the expression levels of research has been performed on this gene until relatively CDCA7 were higher in CRC tissues (75.96%) compared with recently. in adjacent normal colorectal tissues (26.92%; P<0.0001; Fig. 5A). In addition, the RT‑qPCR results indicated Gene‑disease association. Based on the PubMed database, that the expression levels of CDCA7 were significantly the frequency with which CDCA7 was reported to be asso- upregulated among 15 CRC tissues compared to adjacent ciated with disease was collected, and the histogram of the normal colorectal tissues (P=0.044; Fig. 5B). Furthermore, top 20 diseases associated with CDCA7 is shown in Fig. 2. western blotting revealed that similar increased levels of Notably, CDCA7 was most commonly reported in neoplasms, CDCA7 protein were observed in 15 CRC tissues (P=0.016; but its association with CRC has not yet been reported. Hence, Fig. 5C and D). there is still more research that is needed to investigate the expression and significance of CDCA7 in CRC. High expression of CDCA7 in CRC cell lines. CDCA7 expression levels were relatively higher in CRC cell lines compared Regulatory relationships. As shown in Table III, there were to NCM460 (a normal colonic epithelial cell line), as revealed numerous regulatory factors associated with CDCA7, such as by RT‑qPCR (Fig. 6A) and western blotting (Fig. 6B and C) hsa‑let‑7b‑5p and human leukocyte antigen (HLA) complex P5 analyses, suggesting that CDCA7 serves a role in CRC. (HCP5), indicating that this gene may act as a transcriptional regulator. CDCA7 expression is associated with the clinical progression of CRC. The results of RT‑qPCR and western blotting indicated Transcription factor prediction. The transcription factors that CDCA7 expression levels were significantly increased in bound to the promoter region of CDCA7 were predicted based human CRC. To further examine the potential association 62 LI et al: HIGH EXPRESSION OF CDCA7 PREDICTS TUMOR PROGRESSION AND POOR PROGNOSIS IN CRC

Figure 3. Transcription factors of CDCA7. Red color represents high recommendation. CDCA7, cell division cycle associated protein 7.

Figure 4. Expression levels of cell division cycle associated protein 7 in various types of cancer. Red represents tumor tissues, while blue represents normal tissues. between CDCA7 expression and CRC clinicopathological rates of CDCA7 were 26.92% (28/104) and 75.96% (79/104) features, the expression levels of CDCA7 were detected in in normal and CRC tissues, respectively, with a statistically 104 CRC tissues and their paired adjacent non‑tumor tissues significant difference (P<0.0001) (data not shown). CDCA7 via immunohistochemistry staining. The positive expression expression levels in 104 pairs of CRC tissues were catego- Molecular Medicine REPORTS 22: 57-66, 2020 63

Table IV. Association between CDCA7 expression and clinicopathological characteristics of patients with CRC.

CDCA7 Clinicopathological number of ------characteristics cases low High P‑value

Sex 0.533 Male 61 16 45 Female 43 9 34 Age (years) 0.369 <60 58 12 46 ≥60 46 13 33 Tumor size 0.217 <5 cm 64 18 46 ≥5 cm 40 7 33 Invasion depth 0.002 Submucosal 17 9 8 Below the muscle layer 87 16 71 Differentiation 0.318 High 12 3 9 Moderate 78 21 57 Poor 14 1 13 TNM stage I‑II 54 20 34 0.001 III‑IV 50 5 45 Lymph node metastasis 0.038 Negative 65 20 45 Positive 39 5 34 Distant metastasis 0.019 Negative 89 25 64 Positive 15 0 15

CDCA7, cell division cycle associated protein 7; TNM stage, tumor‑node‑metastasis.

rized into high‑expression (n=79) and low‑expression (n=25) groups. As summarized in Table IV, high expression levels of CDCA7 were significantly associated with CRC invasion depth (P=0.002), lymph node metastasis (P=0.038), tumor‑node‑metastasis stage (P=0.001) and distant metastasis (P=0.019). However, no significant differences were found in sex (P=0.533), age (P=0.369), tumor size (P=0.217) and CRC differentiation (P=0.318) between high and low CDCA7 expression groups. Collectively, these results indicated that CDCA7 overexpression may be involved in the progression of CRC. Figure 5. Levels of CDCA7 in CRC tissues. (A) Immunohistochemistry streptavidin peroxidase staining was conducted to determine the levels of High expression of CDCA7 predicts poor prognosis in CDCA7 in CRC tissues and paired adjacent normal colorectal tissues (magnification, x400). (B) Relative mRNA expression levels of CDCA7 in 15 patients with CRC. The association between CDCA7 expres- pairs of CRC tissues and adjacent normal colorectal tissues were assessed by sion and overall survival in patients with CRC was evaluated reverse transcription‑quantitative PCR and (C) western blotting. (D) Relative by Kaplan‑Meier analysis. As shown in Fig. 7, high CDCA7 protein levels of CDCA7 in 15 CRC tissues and their paired adjacent normal expression group was significantly associated with shorter colorectal tissues. GAPDH was used as the internal control. Data were examined by t‑test and are presented as the mean ± SD. CDCA7, cell division cycle survival times as compared with the low CDCA7 expression associated protein 7; CRC, colorectal cancer. group (P=0.012). 64 LI et al: HIGH EXPRESSION OF CDCA7 PREDICTS TUMOR PROGRESSION AND POOR PROGNOSIS IN CRC

Figure 6. Expression levels of CDCA7 mRNA and protein in colorectal cancer cell lines. (A) mRNA expression levels of CDCA7 were detected by fluores- cence‑based quantitative PCR. (B) Western blotting showing the protein expression levels of CDCA7; (C) corresponding histogram. GAPDH was used as the internal control; expression value of CDCA7 was set to 1 in the NCM460 group. The data were analyzed by the one‑way ANOVA and Bonferroni's post hoc test and are presented as the mean ± SD. *P<0.05, **P<0.01 vs. NCM460 group. CDCA7, cell division cycle associated protein 7; N.S., non‑significant.

CDCA7 is a family member of the cell division cycle proteins, which are mainly expressed in the cell nucleus (9). In the present study, the results of immunohistochemical staining demonstrated that CDCA7 was expressed primarily in the nucleus, and at least partly in cytoplasm. A previous study has demonstrated that CDCA7 is a Myc target gene, which participates in Myc‑mediated tumor transformation and eventually leads to tumor occurrence (9). Mitotic stimula- tion can induce the activation of Myc to promote cell cycle progression (23,24). Moreover, CDCA7 is positively expressed in solid tumors, along with high Myc levels (17). These Figure 7. CDCA7 overexpression predicts poor prognosis in patients with findings indicated that CDCA7 may promote tumor growth colorectal cancer. Kaplan‑Meier survival curves were established for both high and low CDCA7 expression groups. High CDCA7 expression progression with Myc. In a previous study, CDCA7 was iden- group (n=79) exhibited a poorer prognosis (P=0.012) compared to the low tified as a novel transcription factor E2F transcription factor 1 CDCA7 expression group (n=25). CDCA7, cell division cycle associated (E2F1)‑induced protein, suggesting that the expression of protein 7. CDCA7 is activated by both E2F1 and Myc (25). Additionally, other studies have shown that Myc and E2F1 can bind to the CDCA7 promoter in order to upregulate the expression levels Discussion of CDCA7 (9,25). These findings suggest that CDCA7 may participate in tumor occurrence via the activation of Myc and CRC is the most common form of gastrointestinal cancer, and E2F1. However, this statement needs further experimental one of the leading causes of cancer‑related mortality glob- verification. ally (21,22). Although clinicopathological parameters have It is worth noting that disordered cell cycle control is been used to predict the clinical outcome of CRC in patients, a basic feature of tumor pathogenesis, and there is a large these classification indicators are not precise enough to predict imbalance in cyclic protein regulation in tumor cells (26‑28). the prognosis of patients with CRC. Therefore, it is essential High CDCA7 expression has been detected in different types to identify novel tumor biomarkers and therapeutic targets for of human cancer, indicating that CDCA7 may function as an early diagnosis and treatment of CRC. This study focused on oncogene. For example, Jiménez‑P et al (14) demonstrated determining the expression levels and clinical role of CDCA7 that CDCA7 protein is upregulated in Burkitt lymphoma in CRC. cell lines and tumor tissues, and CDCA7 mRNA levels are Molecular Medicine REPORTS 22: 57-66, 2020 65 significantly elevated in numerous T and B lymphoma cell Acknowledgements lines. In addition, a previous study reported that CDCA7 is overexpressed in the YDOV‑151 human ovarian cancer cell The authors would like to thank Dr S.Q. Liu, Dr Y.J. Su, line (>7‑fold expression) compared with in human ovarian Mr. T,Y. Zhang and Mr. Y. Zhu at the Department of surface epithelial cells (10). Moreover, CDCA7 has been Gastroenterology, The Second Affiliated Hospital of Guangxi reported to be involved in the occurrence of retinoblastoma, Medical University for their help and excellent technical which can be utilized as a biomarker for early diagnosis and support. treatment of the disease (15). Cheng et al (18) demonstrated that the expression level of CDCA7 is higher in esophageal Funding squamous cell carcinoma compared to normal esophageal tissue. Furthermore, Osthus et al (17) reported that CDCA7 The current study was funded by the National Nature Science is overexpressed in patients with acute myeloid leukemia, Foundation of China (grant no. 81760516) and the Program leading to an increased risk of lymphoid malignancies in for Improvement Scientific Research Ability of Young and these patients. Overall, CDCA7 is upregulated in a wide Middle‑Aged Teachers of Higher Education of Guangxi variety of human tumors, and is likely to be associated with (grant no. 2017KY0093). cancer progression. Additionally, previous evidence has suggested that Availability of data and materials CDCA7 is involved in the proliferation and apoptosis of tumor cells (29). Recently, it has been shown that the lncRNA The datasets used and/or analyzed during the current study are FGD5‑AS1 can promote the proliferation, migration and available from the corresponding author on reasonable request. invasion of CRC cells by upregulating CDCA7 via sponging miR‑302e (30). This study also found that FGD5‑AS1 can Authors' contributions competitively bind with miR‑302e to modulate CDCA7, resulting in the induction of CRC cell apoptosis (30). This CL designed the study and revised the manuscript. MQ and indicated that CDCA7 exhibits a transcriptional regulatory JZ contributed to sample collection and performed the bioin- function and by being modulated by its upstream target formatics analysis. SL and JH performed the experiments, lncRNA, it can affect the progression of CRC. In the present data analysis and wrote the manuscript. All authors read and study, GO term analysis revealed that CDCA7 was related approved the final manuscript. to cell proliferation and apoptosis. While a recent report has focused on the molecular mechanisms of CDCA7 and Ethics approval and consent to participate CRC (30), the exact relationship between CDCA7 and CRC still remains largely unknown. Further research into this Informed consent was obtained from patients. The present study relationship is needed in the future. was approved by the Protection of Human Ethics Committee of The results of the present study indicated that CDCA7 the First Affiliated Hospital of Guangxi Medical University. expression was upregulated in human CRC tissues compared with in adjacent normal tissues. In addition, it was shown Patient consent for publication that high CDCA7 expression could contribute to advanced tumor progression in patients with CRC. The results of Not applicable. Kaplan‑Meier analysis demonstrated that different expression levels of CDCA7 exhibited significant effects on the prognosis Competing interests of patients with CRC (P=0.012). Hence, CDCA7 could be a reliable marker for predicting tumor progression and survival The authors declare that they have no competing interests. prognosis in patients with CRC, but further studies are needed in order to validate this. However, this study is limited by the References lack of clarification on the specific mechanisms underlying the positive association between CDCA7 and CRC progression. 1. Torre LA, Freddie B, Siegel RL, Ferlay J, Lortet‑Tieulent J and Jemal A: Global cancer statistics, 2012. CA Cancer J Clin 65: Therefore, further research and clinical trials into CDCA7 are 87‑108, 2015. needed in the future. 2. Siegel RL, Miller KD and Jemal A: Cancer statistics, 2015. CA In conclusion, the results of the present study provided Cancer J Clin 65: 5‑29, 2015. 3. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, evidence that CDCA7 may be highly expressed in CRC tissues Rebelo M, Parkin DM, Forman D and Bray F: Cancer and may be associated with advanced tumor progression. incidence and mortality worldwide: Sources, methods and Notably, to the best of our knowledge, this is the first study to major patterns in GLOBOCAN 2012. Int J Cancer 136: E359‑E386, 2015. investigate the expression of CDCA7 in CRC tissues and cell 4. Chiu HM, Hsu WF, Chang LC and Wu MH: Colorectal cancer lines, as well as its relationship with the clinical parameters screening in Asia. Curr Gastroenterol Rep 19: 47, 2017. of tumor progression. In addition, these results demonstrated 5. Calon A, Espinet E, Palomo‑Ponce S, Tauriello DV, Iglesias M, Céspedes MV, Sevillano M, Nadal C, Jung P, Zhang XH, et al: that high CDCA7 expression contributes to poor prognosis in Dependency of colorectal cancer on a TGF‑β‑driven program in patients with CRC. Taken together, these results suggested that stromal cells for metastasis initiation. Cancer Cell 22: 571‑584, CDCA7 may serve as a novel biomarker for CRC diagnosis 2012. 6. Cui G, Cai F, Ding Z and Gao L: MMP14 predicts a poor and a reference indicator for the prognosis of patients with prognosis in patients with colorectal cancer. Hum Pathol 83: CRC. 36‑42, 2019. 66 LI et al: HIGH EXPRESSION OF CDCA7 PREDICTS TUMOR PROGRESSION AND POOR PROGNOSIS IN CRC

7. Ogata H, Goto S, Sato K, Fujibuchi W, Bono H and Kanehisa M: 20. Livak KJ and Schmittgen TD: Analysis of relative gene expression KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic data using real‑time quantitative PCR and the 2(‑Delta Delta Acids Res 27: 29‑34, 1999. C(T)) method. Methods 25: 402‑408, 2001. 8. Harris MA, Clark J, Ireland A, Lomax J, Ashburner M, Foulger R, 21. Bray F, Ferlay J, Soerjomataram I, Siegal RL, Torre LA and Eilbeck K, Lewis S, Marshall B, Mungall C, et al: The Gene Jemal A: Global Cancer Statistics 2018: GLOBOCAN Estimates Ontology (GO) database and informatics resource. Nucleic Acids of Incidence and Mortality Worldwide for 36 Cancers in 185 Res 32 (Database Issue): D258‑D261, 2004. Countries, CA Cancer J Clin 68 (6), 394-424, 2018. 9. Prescott JE, Osthus RC, Lee LA, Lewis BC, Shim H, Barrett JF, 22. GBD 2015 Mortality and Causes of Death Collaborators. Global, Guo Q, Hawkins AL, Griffin CA and Dang CV: A novel regional, and national life expectancy, all-cause mortality, and c‑Myc‑responsive gene, JPO1, participates in neoplastic transfor- cause-specific mortality for 249 causes of death, 1980-2015: a mation. J Biol Chem 276: 48276‑48284, 2001. systematic analysis for the Global Burden of Disease Study 2015. 10. Cho H, Lim BJ, Kang ES, Choi JS and Kim JH: Molecular Lancet 388 (10053), 1459-1544, 2016. characterization of a new ovarian cancer cell line, YDOV‑151, 23. Conover CA and Bale LK: Insulin‑like growth factor I induction established from mucinous cystadenocarcinoma. Tohoku J Exp of c‑myc expression in bovine fibroblasts can be blocked by ante- Med 218: 129‑139, 2009. cedent insulin receptor activation. Exp Cell Res 238: 122‑127, 11. Lewis BC, Shim H, Li Q, Wu CS, Lee LA, Maity A and 1998. Dang CV: Identification of putative c‑Myc‑responsive genes: 24. Kitaura H, Shinshi M, Uchikoshi Y, Ono T, Iguchi‑Ariga SM and Characterization of rcl, a novel growth‑related gene. Mol Cell Ariga H: Reciprocal regulation via protein‑protein interaction Biol 17: 4967‑4978, 1997. between c‑Myc and p21(cip1/waf1/sdi1) in DNA replication and 12. Basu Baul TS, Longkumer I, Duthie A, Singh P, Koch B and transcription. J Biol Chem 275: 10477‑10483, 2000. Guedes da Silva MFC: Triphenylstannyl ((arylimino)methyl) 25. Goto Y, Hayashi R, Muramatsu T, Ogawa H, Eguchi I, Oshida Y, benzoates with selective potency that induce G1 and G2/M cell Ohtani K and Yoshida K: JPO1/CDCA7, a novel transcription factor E2F1‑induced protein, possesses intrinsic transcriptional cycle arrest and trigger apoptosis via ROS in human cervical regulator activity. Biochim Biophys Acta 1759: 60‑68, 2006. cancer cells. Dalton Trans 47: 1993‑2008, 2018. 26. Zhang J, Sun Z, Han Y, Yao R, Yue L, Xu Y and Zhang J: Rnf2 13. Wang YC, Chang KC, Lin BW, Lee JC, Lai CH, Lin LJ, Yen Y, knockdown reduces cell viability and promotes cell cycle arrest Lin CS, Yang SJ, Lin PC, et al: The EGF/hnRNP Q1 axis is in gastric cancer cells. Oncol Lett 13: 3817‑3822, 2017. involved in tumorigenesis via the regulation of cell cycle‑related 27. Lee J, Choi BY and Keum YS: Acetonitrile extract of Salvia genes. Exp Mol Med 50: 1‑14: 2018. miltiorrhizaRadix exhibits growth‑inhibitory effects on prostate 14. Jiménez‑P R, Martín‑Cortázar C, Kourani O, Chiodo Y, cancer cells through the induction of cell cycle arrest and Cordoba R, Domínguez‑Franjo MP, Redondo JM, Iglesias T and apoptosis. Oncol Lett 13: 2921‑2928, 2017. Campanero MR: CDCA7 is a critical mediator of lymphoma- 28. Lu J, Chen X, Qu S, Yao B, Xu Y, Wu J, Jin Y and Ma C: Oridonin genesis that selectively regulates anchorage‑independent growth. induces G2/M cell cycle arrest and apoptosis via the PI3K/Akt Haematologica 103: 1669‑1678, 2018. signaling pathway in hormone‑independent prostate cancer cells. 15. Wang QL, Chen X, Zhang MH, Shen QH and Qin ZM: Oncol Lett 13: 2838‑2846, 2017. Identification of hub genes and pathways associated with retino- 29. Gill RM, Gabor TV, Couzens AL and Scheid MP: The blastoma based on co‑expression network analysis. Genet Mol MYC‑associated protein CDCA7 is phosphorylated by AKT to Res 14: 16151‑16161, 2015. regulate MYC‑dependent apoptosis and transformation. Mol 16. Albulescu R: Elevated cyclin B2 expression in invasive breast Cell Biol 33: 498‑513, 2013. carcinoma is associated with unfavorable clinical outcome. BMC 30. Li D, Jiang X, Zhang X, Cao G, Wang D and Chen Z: Long Cancer 3: 1, 2013. noncoding RNA FGD5‑AS1 promotes colorectal cancer cell 17. Osthus RC, Karim B, Prescott JE, Smith BD, McDevitt M, proliferation, migration, and invasion through upregulating Huso DL and Dang CV: The Myc target gene JPO1/CDCA7 CDCA7 via sponging miR‑302e. In Vitro Cell Dev Biol Anim 55: is frequently overexpressed in human tumors and has limited 577‑585, 2019. transforming activity in vivo. Cancer Res 65: 5620‑5627, 2005. 18. Cheng C, Zhou Y, Li H, Xiong T, Li S, Bi Y, Kong P, Wang F, Cui H, Li Y, et al: Whole‑genome sequencing reveals diverse models of structural variations in esophageal squamous cell This work is licensed under a Creative Commons carcinoma. Am J Hum Genet 98: 256‑274, 2016. 19. Tate JG, Bamford S, Jubb HC, Sondka Z, Beare DM, Bindal N, Attribution-NonCommercial-NoDerivatives 4.0 Boutselakis H, Cole CG, Creatore C, Dawson E, et al: COSMIC: International (CC BY-NC-ND 4.0) License. The catalogue of somatic mutations in cancer. Nucleic Acids Res 47: D941‑D947, 2019.